Distributed applications are important for the future of computer networks because they enhance efficiency. Current distributed systems either assume highly available servers or highly available network connections. Coordinating distributed applications is problematic for ad hoc wireless systems because the entire ad hoc network capacity is constrained. Distributed applications are especially problematic to coordinate for smaller, but so-called “smart devices” such as cellular phones, personal digital assistants (PDAs), and laptops that can interact with sensors and controllers and other environmental objects. The connectivity of such devices and objects can be, for example, BlueTooth or IEEE 802.11. Power limitations, radio coverage and the geography distribution of cooperating nodes often prevent an end-to-end path for ad hoc nodes. In general, ad hoc routing methods assume an end-to-end path that connects communication nodes and focus on locating an existing path, such as a series of intermediate nodes connecting cooperating devices. Many applications, such as those requiring real-time communication, require a fully connected path to transfer sequential packets in a timely manner. Other applications do not require a fully connected path and can exist in highly partitioned networks with eventual message delivery. Such applications include personal data/information management, news information propagation and personally messaging. Such applications require only asynchronous messaging service. Known methods provide that if there is no path, communication fails.
The current methodologies for ad hoc networks coordinating distributed applications would benefit from avoidance of requiring end-to-end paths of communication nodes.